IoT botnet attacks detection using deep learning approaches: a review
Vinaykumar Soni, Ashwinikumar Jha · IET conference proceedings. · 2025
The rapid proliferation of the Internet of Things (IoT) has produced major security challenges, particularly with botnet assaults that exploit insecure IoT devices. These risks are always changing and complicated, making it challenging for old security methods such as rule-based or signature-based to recognize and deal with them. Through the application of advanced pattern recognition and anomaly detection techniques, Deep learning models have developed as an effective approach for detecting IoT botnet attacks. This systematic review provides a detailed study about IoT botnet life-cycle stages, various widely available IoT botnet datasets, and a comprehensive examination of various research done for detection using ML and DL techniques. We discuss the recent breakthroughs in IoT security, various botnet datasets, and issues connected with ML and DL-based IoT botnet detection approaches. We also propose potential breakthroughs and future research directions to enhance the resilience and usefulness of deep learning models in defending IoT ecosystems. This evaluation seeks to serve as a beneficial resource for people working in building more robust cybersecurity solutions for IoT networks that address IoT botnet attack detection.